Continuous random distributions

Semester 2021-2022

Course name Instructor’s name: Worksheet created by

Random signals and noise Dr. Montasir Qasymeh Name: Mohammad Tahmid ID: 1076344
Hassan
A continuous random variable is a variable that can take any value within a continuous range of values. There are
various types of continuous random distributions such as the uniform distribution, the exponential distribution, the
gamma distribution, the Weibull distribution and the beta distribution.

The uniform distribution has a probability density function that is positive and nonzero over a single continuous finite
range and zero everywhere else. If this distribution is zero for values less than 𝑎 and values greater than 𝑏, then the non-
zero value taken is determined as follows:

When this distribution is integrated from negative infinity to positive infinity, the result has to equal one. If the non-zero
constant is 𝑚,

𝑚(𝑏 − 𝑎) = 1
1
𝑚=
𝑏−𝑎
This is the nonzero value needed.

To determine the expectation of the uniform distribution,

1 2
𝑏
𝑥 (𝑏 − 𝑎2 ) (𝑏 + 𝑎)
∫ 𝑑𝑥 = 2 =
𝑎 𝑏−𝑎 𝑏−𝑎 2

To determine the variance of the uniform distribution :

1 3
𝑏
𝑥2 (𝑏 − 𝑎3 ) (𝑏3 − 𝑏2 𝑎 + 𝑏2 𝑎 − 𝑏𝑎2 + 𝑏𝑎2 − 𝑎3 ) 𝑏2 + 𝑎𝑏 + 𝑎2
∫ 𝑑𝑥 = 3 = =
𝑎 𝑏−𝑎 𝑏−𝑎 3(𝑏 − 𝑎) 3

𝑏2 + 𝑎𝑏 + 𝑎 2 𝑏 + 𝑎 2 𝑏2 + 𝑎𝑏 + 𝑎2 1 𝑏 + 𝑎 2 4𝑏2 + 4𝑎𝑏 + 4𝑎2 − 3𝑎2 − 6𝑎𝑏 − 3𝑏2 𝑏2 − 2𝑎𝑏 − 𝑎2

−( ) = − ( ) = =
3 2 3 4 1 12 12
(𝑏 − 𝑎)2
=
12
The exponential distribution is as follows:

It has a probability density function as follows:

𝑎𝑒 −𝑎𝑥 𝑤ℎ𝑒𝑛 𝑥 ≥ 0
0 𝑤ℎ𝑒𝑛 𝑥 < 0
Integrating this from minus infinity to positive infinity gives
∞ 0
𝑥=∞
∫ 𝑎𝑒 −𝑎𝑥 𝑑𝑥 + ∫ 0 𝑑𝑥 = [−𝑒 −𝑎𝑥 ]𝑥=0 =1
0 −∞

The expectation of the exponential distribution is as follows:

∞ 0 ∞
1 1
∫ 𝑎𝑥𝑒 −𝑎𝑥 𝑑𝑥 + ∫ 𝑥(0) 𝑑𝑥 = [−𝑥𝑒 −𝑎𝑥 ]∞
0 + ∫ 𝑒 −𝑎𝑥 𝑑𝑥 = 0 − [𝑒 −𝑎𝑥 ]∞
0 =
0 −∞ 0 𝑎 𝑎

To determine the variance of the exponential distribution

∞ 0 ∞
2 2 ∞ −𝑎𝑥 2
∫ 𝑎𝑥 2 𝑒 −𝑎𝑥 𝑑𝑥 + ∫ 𝑥(0) 𝑑𝑥 = −[𝑥 2 𝑒 −𝑎𝑥 ]∞
0 + ∫ 2𝑥𝑒 −𝑎𝑥
𝑑𝑥 = [𝑥 2 −𝑎𝑥 ]∞
𝑒 0 − ( (𝑥𝑒 −𝑎𝑥 )∞
0 + ∫ 𝑒 𝑑𝑥) = 2
0 −∞ 0 𝑎 𝑎 0 𝑎

The variance of this distribution is

2 1 1
2
− 2= 2
𝑎 𝑎 𝑎
There is an interesting property:

The probability of the variable X exceeding a certain positive value m, where X follows the exponential distribution is

𝑒 −𝑎𝑚
Therefore, the probability that X exceeds m + n is

𝑒 −𝑎(𝑚+𝑛)
The probability that X exceeds m + n given that X exceeds m is

𝑒 −𝑎(𝑚+𝑛)
= 𝑒 −𝑎𝑛
𝑒 −𝑎𝑚
So, this probability is the same as the probability that X is greater than n.

The probability distribution of a sum of variables each taking the same exponential distribution is a gamma distribution.

The probability density function for the gamma distribution involving n independent variables with the same distribution
with the same parameter a is

𝑥 𝑛−1 𝑎𝑛 −𝑎𝑥
𝑒 𝑤ℎ𝑒𝑛 𝑥 ≥ 0 𝑎𝑛𝑑 0 𝑤ℎ𝑒𝑛 𝑥 < 0
(𝑛 − 1)!
𝑛
The expectation of the gamma function can be written as 𝑎, as this is a linear sum of variables each following the same
exponential distribution. The variance of this gamma distribution is
𝑛
𝑎2
The Weibull distribution is defined as follows
𝑛
𝑛𝑎(𝑎𝑥)𝑛−1 𝑒 −(𝑎𝑥) 𝑤ℎ𝑒𝑛 𝑥 ≥ 0 𝑎𝑛𝑑 𝑎 > 0 𝑎𝑛𝑑 0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
The expectation can be calculated as follows:
 ∞
𝑛
∫ 𝑎𝑛(𝑎𝑥)𝑛−1 𝑒 −(𝑎𝑥) 𝑑𝑥
0

Let 𝑢 = (𝑎𝑥)𝑛

𝑑𝑢 = 𝑎𝑛(𝑎𝑥)𝑛−1
∞
𝑛
∫ 𝑥𝑎𝑛(𝑎𝑥)𝑛−1 𝑒 −(𝑎𝑥) 𝑑𝑥
0
∞
𝑛
∫ 𝑥𝑎𝑛(𝑎𝑥)𝑛−1 𝑒 −(𝑎𝑥) 𝑑𝑥
0
∞
1 1 1 1
∫ (𝑢)𝑛 𝑒 −𝑢 𝑑𝑢 = ( ) !
0 𝑎 𝑎 𝑛
1 1
Where (𝑛) ! = 𝑛 𝑓𝑎𝑐𝑡𝑜𝑟𝑖𝑎𝑙

The variance is determined as follows:

∞
𝑛
∫ 𝑥 2 𝑎𝑛(𝑎𝑥)𝑛−1 𝑒 −(𝑎𝑥) 𝑑𝑥
0
∞
𝑛
∫ 𝑥 2 𝑎𝑛(𝑎𝑥)𝑛−1 𝑒 −(𝑎𝑥) 𝑑𝑥
0
∞
1 2 1 2
∫ 2
(𝑢)𝑛 𝑒 −𝑢 𝑑𝑢 = 2 ( ) !
0 𝑎 𝑎 𝑛
2 2
Where (𝑛) ! = 𝑛 𝑓𝑎𝑐𝑡𝑜𝑟𝑖𝑎𝑙

The variance is
2
1 2 1 1
2
( ) ! − 2 (( ) !)
𝑎 𝑛 𝑎 𝑛

The beta distribution is as follows:

The probability density function is of the form

𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 ∗ 𝑡 𝛼 (1 − 𝑡)𝛽 𝑓𝑜𝑟 𝑡 > 0, 𝛼 > 0 𝑎𝑛𝑑 𝛽 > 0

The integral of any probability density function, must equal 1.

The factorial function can be written as the following:

∞
𝑛! = ∫ 𝑒 −𝑡 𝑡 𝑛 𝑑𝑡
0

The product of 2 factorials can be expressed as follows:

∞ ∞ ∞ ∞
𝑚! 𝑛! = ∫ 𝑒 −𝑠 𝑠 𝑚 𝑑𝑠 ∫ 𝑒 −𝑡 𝑡 𝑛 𝑑𝑡 = ∫ ∫ 𝑒 −𝑡 𝑡 𝑛 𝑑𝑡 𝑒 −𝑠 𝑠 𝑚 𝑑𝑠
0 0 0 0

Let 𝑠 = 𝑢𝑣 and 𝑡 = 𝑢(1 − 𝑣)

u has to be positive, as it is multiplied with a negative one in the exponent. So, t has to be between 0 and 1.
∞ ∞ ∞ ∞ ∞ 1
𝑚! 𝑛! = ∫ 𝑒 −𝑠 𝑠 𝑚 𝑑𝑠 ∫ 𝑒 −𝑡 𝑡 𝑛 𝑑𝑡 = ∫ ∫ 𝑒 −𝑡 𝑡 𝑛 𝑑𝑡 𝑒 −𝑠 𝑠 𝑚 𝑑𝑠 = ∫ ∫ 𝑒 −𝑢 (𝑢𝑣)𝑛 𝑢𝑚 (1 − 𝑣)𝑚 𝑑𝑣𝑑𝑢
0 0 0 0 0 0
∞ ∞ ∞ ∞ ∞ 1
−𝑠 𝑚 −𝑡 𝑛 −𝑡 𝑛 −𝑠 𝑚
𝑚! 𝑛! = ∫ 𝑒 𝑠 𝑑𝑠 ∫ 𝑒 𝑡 𝑑𝑡 = ∫ ∫ 𝑒 𝑡 𝑑𝑡 𝑒 𝑠 𝑑𝑠 = ∫ ∫ 𝑒 −𝑢 (𝑢𝑣)𝑛 𝑢𝑚 (1 − 𝑣)𝑚 𝑑𝑣𝑑𝑢
0 0 0 0 0 0
1
= (𝑚 + 𝑛)! ∫ (𝑣)𝑛 (1 − 𝑣)𝑚 𝑑𝑣
0

1
𝑚! 𝑛!
= ∫ (𝑣)𝑛 (1 − 𝑣)𝑚 𝑑𝑣
(𝑚 + 𝑛)! 0

(𝑚+𝑛)!
The constant in question is .
𝑚!𝑛!

The expectation for this variable is

(𝑚 + 𝑛)! 1 𝑛+1 (𝑚 + 𝑛)! (𝑛 + 1)! 𝑚! 𝑛+1

∫ (𝑣) (1 − 𝑣)𝑚 𝑑𝑣 = ∗ =
𝑚! 𝑛! 0 𝑚! 𝑛! (𝑚 + 𝑛 + 1)! 𝑚 + 𝑛 + 1

(𝑚 + 𝑛)! 1 𝑛+2 (𝑚 + 𝑛)! (𝑛 + 2)! 𝑚! (𝑛 + 1)(𝑛 + 2)

∫ (𝑣) (1 − 𝑣)𝑚 𝑑𝑣 = ∗ =
𝑚! 𝑛! 0 𝑚! 𝑛! (𝑚 + 𝑛 + 2)! (𝑚 + 𝑛 + 1)(𝑚 + 𝑛 + 2)
2
(𝑛 + 1)(𝑛 + 2) 𝑛+1 𝑛+1 𝑛+2 𝑛+1
−( ) = ( − )
(𝑚 + 𝑛 + 1)(𝑚 + 𝑛 + 2) 𝑚+𝑛+1 𝑚+𝑛+1 𝑚+𝑛+2 𝑚+𝑛+1

This is the variance.

H. Pishro-Nik, Introduction to Probability. Statistics and Random Processes. .

“Weibull distribution,” Wikipedia, 04-Jan-2023. [Online]. Available:

https://en.wikipedia.org/wiki/Weibull_distribution. [Accessed: 19-Jan-2023].

“Beta distribution,” Wikipedia, 06-Jan-2023. [Online]. Available:

https://en.wikipedia.org/wiki/Beta_distribution. [Accessed: 20-Jan-2023].

“Beta function,” Wikipedia, 19-Dec-2022. [Online]. Available: https://en.wikipedia.org/wiki/Beta_function.

[Accessed: 20-Jan-2023].